Recent changes to mpn_get_str/mpn_set_str

[Torbjörn Granlund]

"Marco Bodrato" <bodrato at mail.dm.unipi.it> writes:

  We may also try the division-free algorithm explored by Bouvier-Zimmermann
  in their paper: https://members.loria.fr/PZimmermann/papers/get_str.pdf
  
I found the old code implementing the basecase.  This code is very
likely the basis for the conclusion in the GMP manual where we say this
is not an improvement.  (See the quote in the paper, section 2.2.)

My code is actually from 2002, and back in those days hardware multiply
was not pipelined.  That meant that repeated divrem_1 would be about the
same speed as the corresponding full schoolbook division.

I'll see if I can brush up the code and make use of it for machines with
pipelined multiply.

It is nice that people publish academic papers about my old ideas so
that I am reminded about them.  :-)

-- 
Torbjörn
Please encrypt, key id 0xC8601622